-
As you can see in the title, GPU not recognized. Using the latest drivers for AMD (24.5.1) on a RX7600 XT
``rocm` devices are available.
local-sync devices are available.
local-task devices are a…
-
Now rocm is available on Windows when TensorFlow-rocm will come to Windows?
-
I have two almost identical computer (hardware), with the same GPU: RTX 4000
On one computer(# 1) I got this message right at the begining:
```
import_cuda_impl: initializing gpu module...
/U//.…
-
**To Reproduce**
MGPUSim version of commit ID: c72c904
ROCm 1.7.1
clang-ocl 4.0
Command that recreates the problem
Only modify the matrixmultiplication kernel to:
```bash
#define TILE_SIZE 8
…
-
### What happened?
This error was difficult to pin down to a specific flag, so I thought I would share.
When debugging unet numerics on gfx1100 with WMMA FA etc., I had to try the fx-decomposed at…
-
### Suggestion Description
When we need to do long context inference (using LongLoRA), sometimes errors occur: **need flash-attn version ≥2.1.0**. So wondering if a newer version will follow.
### Op…
-
### Comment:
Is there any appetite to add a ROCm variant? It would be great, particularly since some big HPC like [LUMI](https://docs.lumi-supercomputer.eu/hardware/lumig/) have gone with AMD cards.
…
-
I want to again request AMD support, since it is now much more popular and usable than it has been
-
# Issue
On clicking generate it loads and times out after a while.
# Expected behaviour
It generates a picture using the prompt
# Details
Enfugue v0.1.2 (Linux)
Installed using archive metho…
-
Ubuntu 22.04
Hi everyone
I thought I'd update on this having tried to get ROCm gpu acceleration recognised via onnx.
We created the environment for building GraX as here:
https://github.com/St…